Sun Microsystems, Inc.

Oracle America, Inc.

Kathirgamar Aingaran

Shree Kant

Kenway W Tam

Yuan-Jung D Lin

Chin-Man Kim

Claude R Gauthier

Edgardo F Klass

Sanjay Patel

David J Greenhill

Ramaswamy Sivaramakrishnan

Chaim Nmi Amir

Poonacha P Kongetira

Zhen Wu Liu

Miriam G Blatt

Manjunath D Haritsa

Joydeep Mitra

Lakshminarasimhan Varadadesikan

Hong-Men Su

Hong You

An apparatus and method for scheduling execution of multiple threads on a shared processor resource is described in connection with a multithreaded multiprocessor chip. Using a thread selection policy that switches between available threads every cycle to give priority to the least recently executed or scheduled threads, different threads are able to operate in a way that ensures no deadlocks or livelocks while maximizing aggregate performance and fairness between threads. Prioritization is accomplished by monitoring and sorting thread status information for each thread, including speculative states in which a thread may be speculatively scheduled, thereby improving usage of the execution pipeline by switching a thread in with a lower priority.

Method and apparatus for scheduling multiple threads for execution in a shared microprocessor pipeline

A method for data analysis of power modeling for a microprocessor has been developed. The method takes multiple values of power data from a power modeling simulator and generates summary data to characterize the power data behavior. Summary data views include results characterizing behavior in a single cycle and behavior across multiple cycles. Data is viewed both at an absolute level to characterize total power and relative to previous levels to characterize power derivatives. Summary data is derived from power generated every cycle when running specific benchmark programs on the power simulator.

Data analysis techniques for dynamic power simulation of a CPU

A method and apparatus is disclosed for maintaining coherency between a primary cache and a secondary cache in a directory-based cache system. Upon identifying a parity error in the primary cache, a tag parity packet and a load instruction are sent from the primary cache to the secondary cache. In response to the tag parity packet, each tag entry in the secondary cache that is associated with the parity error is invalidated. Upon receiving an acknowledgment of receipt of the tag parity packet, the primary cache functions to invalidate each tag entry in the primary cache that is associated with the parity error. Then, the secondary cache communicates data requested in the load instruction to the primary cache.

Method and apparatus for primary cache tag error handling

A multi-threaded memory system including a plurality of entries, each one of the plurality of entries including a plurality of threads, each one of the plurality of threads including an active cell and a shared read cell. The shared read cell has an output coupled to a read bit line and a corresponding plurality of inputs coupled to an output of the corresponding active cells in each one of the plurality of threads. A multi-threaded memory system is also described.

Efficient implementation of a read scheme for multi-threaded register file

A memory system includes an active storage circuit and at least one base storage circuit. The at least one base storage circuit is coupled to the active storage circuit though at least one pass gate, at least one driver and a bit line. The at least one pass gate and the at least one driver have a device size substantially similar to a device size of each one of the devices in the active storage circuit and the at least one base storage circuit. A method of swapping data between two storage circuits is also described.

Efficient method of data transfer between register files and memories

Method and apparatus for detecting and analyzing effects of noise in a digital circuit that arises from a coupling of signals produced by switching of a first gate and a second gate in a timed relationship. Where each of a first gate and a second gate can switch within a selected switching time interval, the gate switching effects are combined and the second gate output signal is analyzed with reference to the first gate input signal. Otherwise, the gate switching effects are not combined. When the second gate output signal satisfies at least one of three criteria, this condition is interpreted as indicating that the second gate permits propagation of a noise pulse produced at the first gate.

Windowing scheme for analyzing noise from multiple sources

An automated method of analyzing crosstalk in a digital logic integrated circuit, the method operating on a digital computer, is described. The method uses available software to make an extracted, parameterized netlist from a layout of the integrated circuit. For at least one potential victim wire of the plurality of wires, determining a subset of the wires of the chip are found to be potential aggressor wires that may couple to the victim wire. The aggressor wires are combined into a common aggressor. A risetime of the common aggressor is calculated and used to calculate the magnitude of coupled noise on the victim wire induced by the aggressor wires. An alarm threshold for each potential victim wire is determined based upon the type of logic gate that receives the victim wire. The alarm thresholds for each potential victim wire are compared to the calculated height of a coupled noise on the victim wire to determine which, if any, wires of the design suffer enough crosstalk noise that they should be redesigned.

Two pole coupling noise analysis model for submicron integrated circuit design verification

An automated method of analyzing crosstalk in a digital logic integrated circuit on a digital computer is described. The method uses available software to make an extracted, parameterized netlist from a layout of the integrated circuit. The netlist has gate and black box invocations as well as transistor invocations. Library models are used to find driving resistances and capacitances associated with the gate and black-box invocations. For at least one potential victim wire of the plurality of wires, a subset of the wires of the chip are found to be potential aggressor wires to the victim wire. The aggressor wires are combined into a common aggressor. A risetime of the common aggressor is calculated and used to calculate the magnitude of coupled noise on the victim wire induced by the aggressor wires. An alarm threshold for each potential victim wire is determined based upon the type of logic gate that receives the victim wire. The alarm thresholds for each potential victim wire are compared to the calculated height of a coupled noise on the victim wire to determine which, if any, wires of the design suffer enough crosstalk noise that they should be redesigned.

Method of analyzing crosstalk in a digital logic integrated circuit

Hierarchical coupling noise analysis for submicron integrated circuit designs

A logic gate for producing an output signal representing a logical operation of a first logic signal and a second logic signal includes a first input terminal for receiving the first logic signal and a second input terminal for receiving the second logic signal. The logic gate further includes a first transistor, a second transistor, and an evaluation node which is connected to a pre-charge device. The first transistor has a first terminal coupled to the first input, a second terminal coupled to the evaluation node, and a third terminal coupled to the second input. The second transistor has a first terminal coupled to the second input, a second terminal coupled to the evaluation node, and a third terminal coupled to the first input. A change in either of the logic signals triggers the logic gate, and a change in both of the logic signals within a predetermined time period results in the logic signals simultaneously canceling each other out.

Growing community of inventors

Sunnyvale, CA, United States of America

Kathirgamar Aingaran